CN109242886A - A kind of modeling of space cluster target trajectory and forecasting procedure - Google Patents
A kind of modeling of space cluster target trajectory and forecasting procedure Download PDFInfo
- Publication number
- CN109242886A CN109242886A CN201811035805.4A CN201811035805A CN109242886A CN 109242886 A CN109242886 A CN 109242886A CN 201811035805 A CN201811035805 A CN 201811035805A CN 109242886 A CN109242886 A CN 109242886A
- Authority
- CN
- China
- Prior art keywords
- cluster
- target
- motion
- modeling
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
It is that a kind of space cluster target trajectory is portrayed and forecasting procedure, it is therefore intended that portrayed by the mathematical model of complex target track kinetic characteristic, improve tracking and the forecast precision of such target the present invention relates to aerospace field.Complex target Kinematic Decomposition is well-marked target motion event two parts in cluster mass motion event and cluster by this method.For cluster entirety, its characteristics of motion is portrayed and predicted from cluster centers movement tendency, group space structure evolution and subgroup relative motion respectively；For well-marked target, target is differentiated on the basis of, using layered modeling thought, the technologies such as use in conjunction dynamics, kinematics and compensating Modeling realize portraying and predicting for its kinetic characteristic.This patent the method realizes the modeling of space cluster target trajectory and forecast with the thought combined is locally differentiated using whole tracking, the design that the fields trackings such as space debris environment monitoring, in-orbit service, anti-ballistic early warning and Situation Awareness can be directly served in, improves the tracking performance of complex target.
Description
Technical field
It is a kind of towards complex target, motion trajectory model building and rail the present invention relates to field of aerospace technology
Mark forecasting procedure.
Background technique
Multiple heads of missile attack or bait interference, the collision orbital tracking generated, the panus that not yet scatters and adjoint
Flight all has typical clustering feature at a distance of closer Satellite Formation Flying etc..Complex target, which refers to, is meeting given spacing about
Under the conditions of beam, the relatively-stationary multiple target set in spatial position is kept within the sufficiently long time.Due to heavy dense targets in cluster
And sensor resolution is not enough, target is that part is distinguishable.In addition, interfering with each other, blocking between target, measurement with
Destination number is inconsistent, can not accurate correlation one by one, cause Targets Dots interrupted, it is difficult to stablize and maintain track.Therefore, how to tie up
The stabilization high precision tracking to complex target is held, is target following filtering and the technical problem that related fields needs to solve, space
Target is even more so.
Collective motion is the whole reflection of all target self-movements in group, the mathematical modulo of these movements of succinct accurate description
Type (motion modeling) is to realize more moment metrical information effective uses, improves the important prerequisite of cluster track precision.Existing sky
Between target movement model be mostly for single target, what is utilized is target dynamics characteristic or kinematic parameter variation continuity etc.,
However complex target quantity is more, motion feature is inconsistent, and the absolute movement of complex target center is caused not fully to meet really
Extraterrestrial target motion feature, existing method can not be applied directly in collective motion description.In addition, group space structure and its
The portraying of changing rule, in cluster subgroup the variation of quantity and motion state, it is also desirable under space cluster kinematic constraint, provide
Succinct and accurately description, resolves and is forecast convenient for subsequent cluster motion state.Therefore, it is transported to obtain high-precision complex target
Movable model and track forecast result need to also set about from physical principle and data statistic analysis etc., in mathematical theory and practical technique
It is upper to obtain key breakthrough.
It should be pointed out that for dense distribution and the huge cluster of number of members, by detection energy, time, precision etc.
Limitation, all object detecting and trackings are not necessarily to, it is also possible to cannot achieve.Therefore, in addition to cluster mass motion shape
State description, only need to be high to verification and measurement ratio in cluster or threatens big target to carry out state description, thus realize cluster identification and with
Track.Corresponding is exactly the modeling and forecast of collective motion modeling with well-marked target in forecast and cluster.
Therefore, the present invention fully considers complex target movement characteristic, differentiates using entirety tracking and locally the thought combined,
The modeling and forecast for first carrying out cluster entirety and its structure, can divide in cluster the track of well-marked target carry out individually tracking and
Prediction, to meet the different fields such as space debris environment monitoring, in-orbit service, anti-ballistic early warning and Situation Awareness to space cluster mesh
Target track demand.
Summary of the invention
The technical problem to be solved by the present invention is to the resolvings of space complex target motion profile in Aviation space industry
With forecast demand, a kind of modeling method for portraying collective motion characteristic is provided, conventional method is overcome to model only for simple target
Defect, improve cluster object tracking performance.The basic ideas of technical solution of the present invention are, are collection by complex target Kinematic Decomposition
Well-marked target moves two class events in group's mass motion and cluster, and then establishes motion model appropriate to two class times respectively and retouch
It states and trajectory predictions method.
The technical scheme is that a kind of space cluster target trajectory modeling and forecasting procedure, which is characterized in that
It cluster mass motion is described drawn game part distinguishes to combine, in different ways respectively to well-marked target in cluster entirety and cluster
Movement portrayed, processing method is as follows:
(1) space cluster target motion event decomposes
Collective motion is decomposed into well-marked target in cluster mass motion and cluster and moves two independent sectors；
(2) modeling of cluster mass motion track and forecast
Using cluster as the transformable overall goals of shape, establish cluster mass motion and change in shape model and
Forecasting procedure, specific:
1) using cluster as a virtual target, cluster centers motion modeling is carried out, according to cluster scale and distribution characteristics
Simultaneously computing cluster center is chosen, cluster centers movement mould is established in such a way that dynamics is in conjunction with random process compensating Modeling
Type；
2) according to cluster overall shape change characteristic, the modeling of group space structure is carried out, group space knot is extracted in analysis
Structure feature characterising parameter establishes the mathematical model that parameter is portrayed based on random process；
3) according to the relative position of member in group and length velocity relation, the modeling that subgroup constitutes variation is carried out, space mesh is provided
Mark constitutes the constraining equation of cluster, establishes the relative motion model of subgroup in conjunction with cluster centers movement and resolves model ginseng
Number predicts relative motion and its whether meet to constraint equation, thus provide subgroup generate, be dead, separation, merged event it is pre-
Report；
4) model parameter is resolved according to the model of three above step and actual measurement data, carries out cluster centers and moves rail
Mark, space structure variation, subgroup constitute the forecast changed；
(3) modeling of well-marked target motion profile and forecast
On the basis of well-marked target is effectively identified and separated from cluster, using layered modeling mode, carry out each significant
The motion modeling of target and forecast, specific:
1) Dynamic Modeling of basic exercise ingredient, the spatial position according to locating for target and flying speed feature, analyze it
Physics force-mechanism and main mechanical feature, the kinetic model of basic exercise is established according to Newton's second law；
2) compensating Modeling of order motion feature, or parameter imponderable order motion unintelligible for physics stress at
Point, it analyzes it and changes over time law characteristic, choose the random process for portraying its changing rule or semi-parameter model establishes movement
Learn compensation model；
3) well-marked target motion profile is forecast, resolves model ginseng according to the model of two above step and actual measurement data
Number, and the constraint of cluster entirety Forecast characteristic is combined, carry out the forecast of well-marked target motion profile.
Helpfulness of the invention:
Space cluster target trajectory modeling of the invention and forecasting procedure, there is motion feature to portray comprehensively, method
Realize that simple, model accuracy is high, the strong beneficial effect of real-time task application, specific:
(1) collective motion is decomposed into mass motion and well-marked target moves two parts, cluster mass motion covers again
Cluster centers movement, space structure variation, subgroup constitute the parts such as variation, give comprehensively to the motion feature of complex target
It portrays, meets various space tracing task demands；
(2) dynamics, kinematics and compensating Modeling technology used in modeling have mature technology that can use for reference, press
It can very easily realize according to modeling procedure, and gradually various motor-driven ingredients can be carved by way of layered modeling
It draws, ensure that the requirements for high precision of target trajectory；
(3) space cluster target movement model according to the present invention is provided using recursion mode, and relevant parameter
It can use measurement data online resolution, therefore can use these models and carry out extraterrestrial target real-time tracking and track forecast,
Meets the needs of real-time task.
Detailed description of the invention
The main contents and thinking schematic diagram of Fig. 1 extraterrestrial target motion modeling of the present invention；
Algorithm flow schematic diagram Fig. 2 of the invention；
Specific embodiment
The present invention proposes that a kind of space cluster target trajectory models and forecasting procedure, Fig. 1 give motion modeling
Main contents and thinking, Fig. 2 gives the implementing procedure modeled in the present invention with prediction algorithm, below to the specific implementation of invention
Mode elaborates.
Step 1: space cluster target motion event decomposes
The step refers to based on entirety tracking and locally differentiates the thought combined, and collective motion modeling problem is divided
On the one hand solution guarantees to the comprehensive of collective motion Characterizations, on the other hand that the analysis of each motion feature is independent of one another,
It is simplified the difficulty for establishing model and model structure.Appropriate adjustment constraint condition is needed according to tracing task, is distinguished whole
Space cluster target motion event is decomposed into well-marked target in the movement and cluster of cluster entirety and transported by body and well-marked target classification
Two independent sectors are moved, establish different mathematical models to portray it, and two parts movement is forecast respectively, with synthesis
The complete trajectory predictions of complex target.
Step 2: the modeling of cluster mass motion track and forecast
The step refers on the basis of maneuvering target motion modeling method, establishes cluster Global movement feature and portrays and become
Gesture prediction technique, the filtering for meeting cluster mass motion state resolves and track following indication and threat situation analysis etc. need
It asks.Collective motion includes target relative movement in group center's absolute movement and group, external anti-for target relative movement in group
It should include two parts, first is that the variation of group's space structure, second is that subgroup quantity and the opposite variation constituted in group.
Step 1: cluster centers motion profile modeling
Cluster centers motion modeling is the depicting method for all target's center spatial positions feature that research constitutes cluster, this
When group is integrally regarded as to a virtual target, main body meets the extraterrestrial target characteristics of motion, and needs to consider mesh in group
Mark influence of the relative motion to group center.
Firstly, carrying out the selection of cluster centers according to the actual situation.For radar surveying, usual geometric center tracking
Then it is conducive to large-scale formation to use；When multiple targets member's RCS difference is larger, especially for from the multiple targets of institute's surveillance and tracking and
When, accurately capture Small object when, would be more advantageous using centroid tracking；In the case of about the same for multiple targets member RCS
Under, it is tracked using center of gravity more stable.
Secondly, for the modeling of cluster centers movement.Here main movement feature combination extraterrestrial target motion model adds
To portray, mode of the specific method dynamics in conjunction with compensating Modeling.Since the movement of targets all in cluster all meets space
Target tracks universal law, but influenced by target relative movement, therefore, only consider central body gravitation feature, establishes
The kinetic model of cluster centers main movement ingredient, and then transported to target in remaining order motion composition characteristics and group is opposite
Center variation characteristic caused by dynamic is portrayed, and the compensating Modeling technology based on random process is introduced, to remaining order components into
Row modeling, is such as based on the " method of present statistical model compensating Modeling.
Step 2: group space structure change modeling
Group space structural modeling is the depicting method for studying the features such as spatial form, volume, posture, such as space formation mesh
Target formation change of configuration.
Firstly, providing the differentiating method of group space structure, such as it is based on characteristic point, feature sideline and other characteristic parameters
Shape portray, volume is portrayed, spatial orientation etc., need to this time fully consider and be mentioned in real time by measurement data or motion state valuation
Take these characteristic information feasibilities.
Secondly, being studied by time series analysis, random process and all kinds of motion modeling technologies using these characteristic parameters
Space cluster shape evolution model is established, and the temporal aspect of model key parameter is analyzed.Here mainly foundation
The random process model of each feature, such as Markov model.
Step 3: subgroup quantity and composition variation modeling in cluster
It is the depicting method of the events such as studying subgroup generation, dead, separation, merge that subgroup, which constitutes modeling, in cluster, is closed
Key is the relative positional relationship and relative motion for exploring member in group, and different subgroup generates, is dead, merging and separating etc.
Model.
Firstly, the condition met needed for analysis space target configuration cluster, such as opposite cohesion, heading be roughly the same,
There are certain intervals etc., establish the constraint equation of these conditions, and then combine the cluster centers equation of motion, according to Physical Mechanism,
Accompanying flying target is established based on kinematics, bionics, mathematical function, random process, stochastic differential equation or subgroup relative position is closed
The mathematical model of system and relative motion.
Secondly, research subgroup generates, is dead, separation, merge etc. during target Aggregation Characteristics and relative position, speed in group
The changing rule of degree etc. establishes the set description of subgroup number of variations, and then utilizes the methods of graph theory and Bayes's evolved network,
The equation of motion for describing these processes is established, while studying the judgment condition of group number and structure change.
Step 4: cluster mass motion forecast
It herein include the forecast of cluster centers track and its space structure variation forecast two parts.
Forecast for cluster centers, when simple process can be equivalent to the track forecast an of virtual target, but for
Steady high precision tracking situation, is according to tactical backgrounds, target property and radar function first, selection can represent cluster position
It sets and the relatively stable cluster centers of spatial position and velocity variations, such as common geometric center, mass center and center of gravity, and explores collection
The improvement of group center's calculation method；Followed by suitable location prediction mould is selected with reference to the changing rule of space cluster centrode
Type is carried out model parameter On-line Estimation according to measured data, and then is carried out the forecast of center flight path using these models.Such as
Preliminary analysis shows that geometric center is tracked suitable for large-scale formation, and mass center favorably accurately captures Small object, center of gravity then for
Target following RCS about the same is more stable.
For the space structure of cluster, firstly, predictable typical space shape, body in group space structure are extracted in analysis
Product, feature sideline and characteristic point carry out the classification of long-term forecasting and short-period forecast feature according to different purposes；Secondly, according to upper
Model and actual measurement data are stated, model parameter On-line Estimation is carried out, and then carries out the space structure of cluster using these models
Forecast；Finally, according to collective motion speed and relative motion estimated result, especially structure size change rate etc., prediction can
The events such as the group's generation of subgroup existing for energy, dead, separation, merging.
What it is due to this method foundation is using the recursion motion model of state-space method description, according to measurement
Non-linear property carries out the On-line Estimation of model parameter using no mark filtering (the unscented filter, UKF) here.Mould
Type forecast uses numerical integration method, can take into account computational efficiency and the suitable integration method of accuracy selection.
Step 3: the modeling of well-marked target motion profile and forecast
The step refers to well-marked target on the basis of effectively identifying and separating in cluster, using layered modeling thought,
Motion modeling and the forecast of each well-marked target are carried out, is met pre- to these target states filtering resolving and track following
Show and the demands such as threat situation analysis.Tracked and forecast for target is paid close attention in cluster, both met highest priority with
Track needs, but also as the useful supplement of clustering feature forecast.
Step 1: the Dynamic Modeling of basic exercise ingredient
Here modeling is described primarily directed to the kinetic characteristic of single extraterrestrial target, it is intended to pass through motion feature ingredient
It gradually decomposes and extracts, realize that the high-precision of motion state is portrayed.In view of extraterrestrial target movement characteristic, main movement rule can
To be obtained by the physical analysis of force-mechanism, to establish motion model according to Newton's second law.This method mainly for
Terrestrial space target establishes motion model.The demand of real-time tracking and forecast is taken into account, used power model mainly has earth matter
Point gravitation, the aspherical gravitation of the earth (20 × 20), life particle gravitation, atmospheric drag and solar light pressure (fixed area-mass ratio), mould
Type is established under the Celestial Reference System of the earth's core.
Step 2: the compensating Modeling of order motion feature
On the basis of main movement character separation, further playing different models, to portray different motion feature capabilities excellent
Gesture introduces the compensating Modeling method based on random process parameter identification, the order motion ingredient of Semi-parametric estimate technology, establishes base
In compensating Modeling method.The final adaptive motion modeling realized based on co-ordinative construction.
Step 3: well-marked target motion profile forecast
Cluster centers are only the probability positions for indicating target group, serve space situation awareness etc., but for certain spies
Determine task, more concerned be well-marked target in cluster movement, as trajectory intercept in incoming warhead, there is collision in panus
The fragment of risk.Therefore, also need to establish the track forecasting procedure of well-marked target.
This is built upon well-marked target and effectively identifies and separate from cluster, and motion state valuation has been equipped with certain essence
On the basis of degree.At this point, the track forecast of well-marked target is exactly typical extraterrestrial target track forecast, it can be in aforesaid space
On the basis of target motion modeling, in conjunction with real time kinematics state estimation result, suitable motion model is selected to be forecast.This
Outside, the forecast of group space distribution, the especially forecast result of cluster centers can be forecast to provide important for well-marked target track
Reference.
Ibid, due to this method establish be using state-space method description recursion motion model, basis
The non-linear property of measurement carries out estimating online for model parameter using no mark filtering (the unscented filter, UKF) here
Meter, model prediction use numerical integration method.
Claims (7)
1. a kind of space cluster target trajectory modeling and forecasting procedure, which comprises the following steps:
(1) complex target motion modeling is decomposed into well-marked target motion event two in cluster mass motion event and cluster
Point；
(2) mathematical model is carried out to cluster mass motion characteristic to portray, and provide motion profile prediction technique；
(3) mathematical model is carried out to well-marked target in cluster to portray, and provide motion profile prediction technique.
2. the method according to claim 1, wherein according to tracing task needs, will collect in the step (1)
Group's Kinematic Decomposition is that well-marked target moves two independent sectors in the movement and cluster of cluster entirety.
3. the method according to claim 1, wherein in the step (2), firstly, integrally regarding cluster as one
A virtual target carries out cluster centers motion modeling；Secondly, carrying out group space knot according to cluster overall shape change characteristic
The modeling of structure；Again, according to the relative position of member in group and length velocity relation, the modeling that subgroup constitutes variation is carried out；Finally, root
Model parameter is resolved according to above-mentioned model and actual measurement data, and carries out the motion profile forecast of cluster entirety.
4. the method according to claim 1, wherein in the step (3), well-marked target from cluster effectively
On the basis of identifying and separating, by the way of layered modeling, moved into firstly, for known to target stress Physical Mechanism
Point, using Dynamic Modeling；Secondly, being portrayed for remaining order components using compensating Modeling technology；Finally, according to above-mentioned mould
Type and actual measurement data resolve model parameter, carry out the motion profile forecast of each target.
5. according to the method described in claim 3, it is characterized in that, in the step (2), cluster centers motion model by with
Lower method is established: according to cluster scale and distribution characteristics chooses and computing cluster center, is compensated using dynamics and random process
The mode that modeling combines establishes cluster centers motion model.
6. according to the method described in claim 3, it is characterized in that, the model of group space structure passes through in the step (2)
Following methods are established: group space structure feature characterising parameter is extracted in analysis, establishes the mathematics that parameter is portrayed based on random process
Model.
7. according to the method described in claim 3, it is characterized in that, the subgroup of cluster constitutes and structure is drilled in the step (2)
Change model to establish by the following method: providing the constraining equation that extraterrestrial target constitutes cluster, is built in conjunction with cluster centers movement
Whether the relative motion model of vertical subgroup simultaneously resolves model parameter, predict relative motion and its meet to constraint equation, to build
Vertical subgroup generates, is dead, separating and combined model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811035805.4A CN109242886B (en) | 2018-09-06 | 2018-09-06 | Space cluster target motion trajectory modeling and forecasting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811035805.4A CN109242886B (en) | 2018-09-06 | 2018-09-06 | Space cluster target motion trajectory modeling and forecasting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109242886A true CN109242886A (en) | 2019-01-18 |
| CN109242886B CN109242886B (en) | 2021-03-12 |
Family
ID=65060788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811035805.4A Active CN109242886B (en) | 2018-09-06 | 2018-09-06 | Space cluster target motion trajectory modeling and forecasting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109242886B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807790A (en) * | 2019-10-31 | 2020-02-18 | 智慧视通(杭州)科技发展有限公司 | Image data extraction and compression method for video target trajectory tracking content |
| CN110917619A (en) * | 2019-11-18 | 2020-03-27 | 腾讯科技（深圳）有限公司 | Interactive property control method, device, terminal and storage medium |
| CN111784738A (en) * | 2020-06-19 | 2020-10-16 | 中国科学院国家空间科学中心 | Extremely dark and weak moving target correlation detection method based on fluctuation analysis |
| CN109242886B (en) * | 2018-09-06 | 2021-03-12 | 中国人民解放军63921部队 | Space cluster target motion trajectory modeling and forecasting method |
| CN112668652A (en) * | 2020-12-31 | 2021-04-16 | 哈尔滨工业大学 | Method and system for identifying cluster array and motion trend in unmanned equipment confrontation |
| CN113449838A (en) * | 2021-07-05 | 2021-09-28 | 中国人民解放军国防科技大学 | Biological particle cluster construction method based on BCCA optimization model |
| CN113702940A (en) * | 2021-09-18 | 2021-11-26 | 中国人民解放军63921部队 | Spatial cluster target resolution method based on multi-element characteristic information hierarchical fusion and application |
| CN114169066A (en) * | 2021-09-18 | 2022-03-11 | 中国人民解放军63921部队 | Space target characteristic measuring and reconnaissance method based on micro-nano constellation approaching reconnaissance |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156880A (en) * | 2011-04-11 | 2011-08-17 | 上海交通大学 | Method for detecting abnormal crowd behavior based on improved social force model |
| CN106872971A (en) * | 2017-03-16 | 2017-06-20 | 中国民航科学技术研究院 | A kind of flying bird multiple targets tracking based on Swarm Intelligent Model |
| CN106873628A (en) * | 2017-04-12 | 2017-06-20 | 北京理工大学 | A kind of multiple no-manned plane tracks the collaboration paths planning method of many maneuvering targets |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242886B (en) * | 2018-09-06 | 2021-03-12 | 中国人民解放军63921部队 | Space cluster target motion trajectory modeling and forecasting method |
-
2018
- 2018-09-06 CN CN201811035805.4A patent/CN109242886B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156880A (en) * | 2011-04-11 | 2011-08-17 | 上海交通大学 | Method for detecting abnormal crowd behavior based on improved social force model |
| CN106872971A (en) * | 2017-03-16 | 2017-06-20 | 中国民航科学技术研究院 | A kind of flying bird multiple targets tracking based on Swarm Intelligent Model |
| CN106873628A (en) * | 2017-04-12 | 2017-06-20 | 北京理工大学 | A kind of multiple no-manned plane tracks the collaboration paths planning method of many maneuvering targets |
Non-Patent Citations (4)
| Title |
|---|
| CYNTHIA SUNG等: "Trajectory Clustering for Motion Prediction", 《2012 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS》 * |
| 孟鹏飞: "再入目标的跟踪定位与轨迹预报研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 翟光 等: "基于集群空间机器人的合作目标", 《北京理工大学学报》 * |
| 郭荣华 等: "一种基于编队中心的目标跟踪算法", 《新型工业化》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242886B (en) * | 2018-09-06 | 2021-03-12 | 中国人民解放军63921部队 | Space cluster target motion trajectory modeling and forecasting method |
| CN110807790A (en) * | 2019-10-31 | 2020-02-18 | 智慧视通(杭州)科技发展有限公司 | Image data extraction and compression method for video target trajectory tracking content |
| CN110807790B (en) * | 2019-10-31 | 2022-06-03 | 智慧视通(杭州)科技发展有限公司 | Image data extraction and compression method for video target trajectory tracking content |
| CN110917619B (en) * | 2019-11-18 | 2020-12-25 | 腾讯科技（深圳）有限公司 | Interactive property control method, device, terminal and storage medium |
| CN110917619A (en) * | 2019-11-18 | 2020-03-27 | 腾讯科技（深圳）有限公司 | Interactive property control method, device, terminal and storage medium |
| CN111784738A (en) * | 2020-06-19 | 2020-10-16 | 中国科学院国家空间科学中心 | Extremely dark and weak moving target correlation detection method based on fluctuation analysis |
| CN111784738B (en) * | 2020-06-19 | 2023-10-31 | 中国科学院国家空间科学中心 | Extremely dark and weak moving target association detection method based on fluctuation analysis |
| CN112668652A (en) * | 2020-12-31 | 2021-04-16 | 哈尔滨工业大学 | Method and system for identifying cluster array and motion trend in unmanned equipment confrontation |
| CN113449838A (en) * | 2021-07-05 | 2021-09-28 | 中国人民解放军国防科技大学 | Biological particle cluster construction method based on BCCA optimization model |
| CN113449838B (en) * | 2021-07-05 | 2022-06-17 | 中国人民解放军国防科技大学 | Biological particle cluster construction method based on BCCA optimization model |
| CN113702940A (en) * | 2021-09-18 | 2021-11-26 | 中国人民解放军63921部队 | Spatial cluster target resolution method based on multi-element characteristic information hierarchical fusion and application |
| CN114169066A (en) * | 2021-09-18 | 2022-03-11 | 中国人民解放军63921部队 | Space target characteristic measuring and reconnaissance method based on micro-nano constellation approaching reconnaissance |
| CN114169066B (en) * | 2021-09-18 | 2022-07-29 | 中国人民解放军63921部队 | Space target characteristic measuring and reconnaissance method based on micro-nano constellation approaching reconnaissance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109242886B (en) | 2021-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109242886A (en) | A kind of modeling of space cluster target trajectory and forecasting procedure | |
| Lin et al. | An algorithm for trajectory prediction of flight plan based on relative motion between positions | |
| Zeng et al. | A deep learning approach for aircraft trajectory prediction in terminal airspace | |
| CN108764560A (en) | Aircraft scene trajectory predictions method based on shot and long term Memory Neural Networks | |
| CN110018453B (en) | Intelligent model identification method based on aircraft track characteristics | |
| Lawrance et al. | Path planning for autonomous soaring flight in dynamic wind fields | |
| CN104730537B (en) | Infrared/laser radar data fusion target tracking method based on multi-scale model | |
| CN105301584B (en) | The IPPHDF multiple-moving target tracking methods of fuzzy distance solution simultaneously | |
| Xie et al. | Dual-channel and bidirectional neural network for hypersonic glide vehicle trajectory prediction | |
| CN113868971A (en) | Airport area three-dimensional refined wind field reconstruction method based on numerical simulation model and historical wind field characteristics | |
| Lin et al. | Approach for 4-d trajectory management based on HMM and trajectory similarity | |
| CN112862171A (en) | Flight arrival time prediction method based on space-time neural network | |
| Schimpf et al. | Flight trajectory prediction based on hybrid-recurrent networks | |
| Zhang et al. | Predicting tropical cyclogenesis using a deep learning method from gridded satellite and ERA5 reanalysis data in the Western North Pacific basin | |
| Jiang et al. | Research on method of trajectory prediction in aircraft flight based on aircraft performance and historical track data | |
| Kant et al. | Long short-term memory auto-encoder-based position prediction model for fixed-wing uav during communication failure | |
| Gao et al. | A spatio-temporal neural network for fine-scale wind field nowcasting based on lidar observation | |
| CN116029130B (en) | Commercial aerospace accident risk level assessment method | |
| Jiang et al. | Behavior pattern mining based on spatiotemporal trajectory multidimensional information fusion | |
| CN114740467B (en) | Cluster target tracking and quantity and contour dynamic estimation method based on amplitude point trace | |
| Wang et al. | A flight trajectory prediction method based on trajectory clustering | |
| Fernandes et al. | Prediction of critical aircraft performance model parameters from historical flight data | |
| Gan et al. | Research on vehicle trajectory prediction method for intersections without signal lights | |
| Yang et al. | Study on the trajectory positioning and prediction framework of unmanned aerial vehicle based on long short-term memory neural network | |
| Rohani et al. | Machine Learning Approach for Aircraft Performance Model Parameter Estimation for Trajectory Prediction Applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |